Harlosh DCNN0530 – An example of the changing profile of Met Office weather stations.

57.387083, -6.519048 Met Office CIMO Assessed Class 4 Installed 6/4/2023

I doubt many readers will know where Harlosh is – I certainly had never heard of it until starting this project. It is a tiny hamlet of few properties on a peninsula of the same name on the mid western coast of the Isle of Skye. The very new weather station is manually observed, only reporting a limited range of parameters and considered part of the volunteer network of “Climate” reporting stations. It is also the only Met Office site left on the Isle of Skye and represents a significant ongoing change in the style their network.

There are official records from weather stations on Skye dating back to 1859 (Portree) and it is likely that stations well predated even that. For shipping and forecasting, Skye (in common with all the major Scottish Isles) was a key area to assemble data. A quick scan through historic records shows over a dozen sites have existed at various times on the island with Portree running up to 1977 dominating the series along with Prabost from 1959 to 1997.

Taking over reporting for the island to account for when Prabost closed was the station at Lusa on the A87 near the Broadford Aerodrome on the extreme eastern side of the island and close by the ferry terminal to Kyle of Lochalsh. Installed in 1995 the Lusa site was a typical latter 20th Century fully equipped one playing a key role in national weather forecasting as seen below from the roadside camera on Google Streetview.

However, this site was dismantled in 2017 for reasons I have not been able to verify, though a potential land sale in the area is a possible cause. Determining cold weather payments on the island moved to using data from the mainland Achnagart site – how representative of the Isle of Skye this site may actually be is rather questionable.

For the first time in over 150 years there was no official weather reporting site on the island. This marks a notable shift in Met Office strategy who now clearly state that “satellites now supply the majority of observations assimilated into forecast models, particularly valuable over oceans and remote regions” This emphasis change is reflected in many site closures in recent times. For example, the late Tim Channon reviewed all the World Meteorological Organisation reporting sites back in 2011/12 – since then 15 (10%) of these have since closed.

The timeline of these closures is interesting. The peak number of temperature recording sites was reached in 1992 with over 630 observation points. A (slow at first) decline occurred with still 600 sites at the turn of the millenium before a rapid reduction. In the following 10 years alone over 100 sites were closed eventually dropping down to the approximate 390 of 2026. – the same number as in the 1950s.

{Image credit: https://rmets.onlinelibrary.wiley.com/doi/10.1002/gdj3.78}

A large proportion of these closures were caused by the requirement for automation of sites to improve speed of access to data for forecasting purposes. Thus the proportion of instant reading Platinum Resistance Thermometers (PRT) to Liquid in Glass (LIGT) increased and multuiple daily readings similarly increased up until 2017 when LIGTs effectively decame redundant for maximum readings only being retained at manually observed sites. There are even now questions whether these LIGT minimum units are even used by many observers, a view which the Met Office themselves reinforced by quoting timed minima from manual sites as at Shirburn. These new instruments also changed the recording protocols and largely eliminated carry over effects of previous day’s extremes from the traditional systems.

A question a sceptical observer may ask is whether the graph below’s almost mirror image of that above from 1990 onward in any reflects these changes.

The relevance of all this to the new Harlosh site is that rather than a sophisticated automatically reporting full site, the modern variant is an amateur manually reporting unenclosed site with a poorly rated exposure that sits in completely the oppsite of the island to its predecessors in a distinctly different micro-climate by a sea inlet. This location is unlikely to even represent much of Skye itself let alone anywhere else. Frustratingly I could not find any ground view images as the site is so new – if anyone an it would be most useful.

Such sites operate largely on trust rather than any scientific rigour. It is highly unlikely anyone is going to pitch a tent inside the former Lusa enclosure but as was demonstrated at Derrylin such activity can readily occur in an unregulated back garden.

Image courtesy google Streetview:Derrylin with screen almost invisible in front of tents to the right of its solar panel power supply.

There are vastly too many modern Met Office sites in such substandard/ low grade locations prone to external influences and the Met office seems perfectly willing to add more amd more as times goes by. Walled Gardens at Bisterne and Kingston Maurward, alongside poly tunnels such as Bewcastle , and indeed almost every new site is falling into the lowest possible CIMO categories -rather than improving the network it is getting drastically worse.

I had formerly decided to not bother covering every single operational Met office weather station as I did not feel some of them were really worth considering i.e. Cairngorm Summit at 1237 metres amsl or Bealach-na-Ba at 773 metres (despite its spectacular road!). On reflection I have decided that completion is important and thus I will cover all sites and still have almost 20 operational stations to go. My next report leads on from this Harlosh site to Raasay an island close to Skye of just 140 residents. Quite why the Met office felt the necessity to adopt this one also in 2023 is well worth the investigation.

Source: https://talkshop.blog/2026/10/10/harlosh-dcnn0530-an-example-of-the-changing-profile-of-met-office-weather-stations/